Lung Cancer Prediction using Extended KNN Algorithm

Abstract

Among several different types of cancer the one that causes high mortality in every country is lung carcinoma. The possibility of survival from this deadly disease can be enhanced by identifying cancer at an early stage. This paper focuses on an Extended version of the KNN Algorithm that is used for the prediction of lung carcinoma based on the Computed Tomography (CT) - Images given as the input. The 2-D image undergoes a Modified Gabor Filtration technique wherein the images are used to extract the features for Edge Detection. This further undergoes Feature Extraction followed by Binarization which is fed as Production data to the Machine Learning model. Based on the Extended KNN Algorithm, the model evaluates the testing data and corresponding predictions are made. The model predicts the Cancer Stage based on the input CT - Image which is passed to the doctor for further medication.